Synergizing Random Forest and K Means Algorithms: An Analytical Study for Precise Crop Recommendation in Southeast Asia

Abstract

Crop detection and classification are pivotal for optimizing agricultural practices and ensuring sustainable farming. This research presents a sophisticated approach to identifying optimal environments for various crops using advanced machine-learning techniques. The study employs a Random Forest classifier framework to categorize crops based on crucial environmental parameters, including soil nitrogen, phosphorus, potassium levels, temperature, humidity, soil pH, and rainfall. Additionally, a K-Means clustering algorithm groups crops with similar growth conditions. The model demonstrates superior performance compared to existing state-of-the-art approaches, achieving an accuracy of 0.97 and macro average scores of 0.94 for precision, 0.95 for recall, and 0.94 for F1-score. Findings underscore distinct environmental requirements for different crop groups, such as those thriving in arid conditions with minimal rainfall and nutrient content, versus those favoring humid conditions with abundant rainfall and nutrient richness. This study emphasizes the potential of machine learning models to enhance agricultural productivity by aligning crop selection with suitable environmental conditions, facilitating precise agricultural decision-making. The high accuracy and detailed classification underscore the model's efficacy in identifying optimal crop environments, which can significantly improve crop yield and resource management.